以wifi-example-sim.cc爲例說明NS3統計數據模型

利用NS3已有的Trace系統或者Log機制收集記錄和統計數據，例如MAC層收發幀數目，網絡層以上收發包數目的跟蹤與統計，這裏選取example/stats/wifi-example-sim.cc爲例來很好說明問題：

這個仿真程序是一個簡單的實驗，包括兩個節點，基於AdhocMAC信道模型，包含NS3仿真所需常見模型如節點/網絡設備/協議棧和應用進程，這裏的應用進程Sender 和Receiver，基於UDP的不可靠連接。

/* -*- Mode:C++; c-file-style:"gnu"; indent-tabs-mode:nil; -*- */
/*
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation;
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 *
 * Authors: Joe Kopena <[email protected]>
 *
 * This program conducts a simple experiment: It places two nodes at a
 * parameterized distance apart.  One node generates packets and the
 * other node receives.  The stat framework collects data on packet
 * loss.  Outside of this program, a control script uses that data to
 * produce graphs presenting performance at the varying distances.
 * This isn't a typical simulation but is a common "experiment"
 * performed in real life and serves as an accessible exemplar for the
 * stat framework.  It also gives some intuition on the behavior and
 * basic reasonability of the NS-3 WiFi models.
 *
 * Applications used by this program are in test02-apps.h and
 * test02-apps.cc, which should be in the same place as this file.
 * 
 */

#include <ctime>

#include <sstream>

#include "ns3/core-module.h"
#include "ns3/network-module.h"
#include "ns3/mobility-module.h"
#include "ns3/wifi-module.h"
#include "ns3/internet-module.h"

#include "ns3/stats-module.h"

#include "wifi-example-apps.h"

using namespace ns3;
using namespace std;

NS_LOG_COMPONENT_DEFINE ("WiFiDistanceExperiment");

void TxCallback (Ptr<CounterCalculator<uint32_t> > datac,
                 std::string path, Ptr<const Packet> packet) {
  NS_LOG_INFO ("Sent frame counted in " <<
               datac->GetKey ());
  datac->Update ();
  // end TxCallback
}




//----------------------------------------------------------------------
//-- main
//----------------------------------------------
int main (int argc, char *argv[]) {

  double distance = 50.0;
  string format ("omnet");

  string experiment ("wifi-distance-test");
  string strategy ("wifi-default");
  string input;
  string runID;

  {
    stringstream sstr;
    sstr << "run-" << time (NULL);
    runID = sstr.str ();
  }

  // Set up command line parameters used to control the experiment.
  CommandLine cmd;
  cmd.AddValue ("distance", "Distance apart to place nodes (in meters).",
                distance);
  cmd.AddValue ("format", "Format to use for data output.",
                format);
  cmd.AddValue ("experiment", "Identifier for experiment.",
                experiment);
  cmd.AddValue ("strategy", "Identifier for strategy.",
                strategy);
  cmd.AddValue ("run", "Identifier for run.",
                runID);
  cmd.Parse (argc, argv);

  if (format != "omnet" && format != "db") {
      NS_LOG_ERROR ("Unknown output format '" << format << "'");
      return -1;
    }

  #ifndef STATS_HAS_SQLITE3
  if (format == "db") {
      NS_LOG_ERROR ("sqlite support not compiled in.");
      return -1;
    }
  #endif

  {
    stringstream sstr ("");
    sstr << distance;
    input = sstr.str ();
  }




  //------------------------------------------------------------
  //-- Create nodes and network stacks
  //--------------------------------------------
  NS_LOG_INFO ("Creating nodes.");
  NodeContainer nodes;
  nodes.Create (2);

  NS_LOG_INFO ("Installing WiFi and Internet stack.");
  WifiHelper wifi;
  WifiMacHelper wifiMac;
  wifiMac.SetType ("ns3::AdhocWifiMac");
  YansWifiPhyHelper wifiPhy = YansWifiPhyHelper::Default ();
  YansWifiChannelHelper wifiChannel = YansWifiChannelHelper::Default ();
  wifiPhy.SetChannel (wifiChannel.Create ());
  NetDeviceContainer nodeDevices = wifi.Install (wifiPhy, wifiMac, nodes);

  InternetStackHelper internet;
  internet.Install (nodes);
  Ipv4AddressHelper ipAddrs;
  ipAddrs.SetBase ("192.168.0.0", "255.255.255.0");
  ipAddrs.Assign (nodeDevices);




  //------------------------------------------------------------
  //-- Setup physical layout
  //--------------------------------------------
  NS_LOG_INFO ("Installing static mobility; distance " << distance << " .");
  MobilityHelper mobility;
  Ptr<ListPositionAllocator> positionAlloc =
    CreateObject<ListPositionAllocator>();
  positionAlloc->Add (Vector (0.0, 0.0, 0.0));
  positionAlloc->Add (Vector (0.0, distance, 0.0));
  mobility.SetPositionAllocator (positionAlloc);
  mobility.Install (nodes);




  //------------------------------------------------------------
  //-- Create a custom traffic source and sink
  //--------------------------------------------
  NS_LOG_INFO ("Create traffic source & sink.");
  Ptr<Node> appSource = NodeList::GetNode (0);
  Ptr<Sender> sender = CreateObject<Sender>();
  appSource->AddApplication (sender);
  sender->SetStartTime (Seconds (1));

  Ptr<Node> appSink = NodeList::GetNode (1);
  Ptr<Receiver> receiver = CreateObject<Receiver>();
  appSink->AddApplication (receiver);
  receiver->SetStartTime (Seconds (0));

  Config::Set ("/NodeList/*/ApplicationList/*/$Sender/Destination",
               Ipv4AddressValue ("192.168.0.2"));




  //------------------------------------------------------------
  //-- Setup stats and data collection
  //--------------------------------------------

  // Create a DataCollector object to hold information about this run.
  DataCollector data;
  data.DescribeRun (experiment,
                    strategy,
                    input,
                    runID);

  // Add any information we wish to record about this run.
  data.AddMetadata ("author", "tjkopena");


  // Create a counter to track how many frames are generated.  Updates
  // are triggered by the trace signal generated by the WiFi MAC model
  // object.  Here we connect the counter to the signal via the simple
  // TxCallback() glue function defined above.
  Ptr<CounterCalculator<uint32_t> > totalTx =
    CreateObject<CounterCalculator<uint32_t> >();
  totalTx->SetKey ("wifi-tx-frames");
  totalTx->SetContext ("node[0]");
  Config::Connect ("/NodeList/0/DeviceList/*/$ns3::WifiNetDevice/Mac/MacTx",
                   MakeBoundCallback (&TxCallback, totalTx));
  data.AddDataCalculator (totalTx);

  // This is similar, but creates a counter to track how many frames
  // are received.  Instead of our own glue function, this uses a
  // method of an adapter class to connect a counter directly to the
  // trace signal generated by the WiFi MAC.
  Ptr<PacketCounterCalculator> totalRx =
    CreateObject<PacketCounterCalculator>();
  totalRx->SetKey ("wifi-rx-frames");
  totalRx->SetContext ("node[1]");
  Config::Connect ("/NodeList/1/DeviceList/*/$ns3::WifiNetDevice/Mac/MacRx",
                   MakeCallback (&PacketCounterCalculator::PacketUpdate,
                                 totalRx));
  data.AddDataCalculator (totalRx);




  // This counter tracks how many packets---as opposed to frames---are
  // generated.  This is connected directly to a trace signal provided
  // by our Sender class.
  Ptr<PacketCounterCalculator> appTx =
    CreateObject<PacketCounterCalculator>();
  appTx->SetKey ("sender-tx-packets");
  appTx->SetContext ("node[0]");
  Config::Connect ("/NodeList/0/ApplicationList/*/$Sender/Tx",
                   MakeCallback (&PacketCounterCalculator::PacketUpdate,
                                 appTx));
  data.AddDataCalculator (appTx);

  // Here a counter for received packets is directly manipulated by
  // one of the custom objects in our simulation, the Receiver
  // Application.  The Receiver object is given a pointer to the
  // counter and calls its Update() method whenever a packet arrives.
  Ptr<CounterCalculator<> > appRx =
    CreateObject<CounterCalculator<> >();
  appRx->SetKey ("receiver-rx-packets");
  appRx->SetContext ("node[1]");
  receiver->SetCounter (appRx);
  data.AddDataCalculator (appRx);




  /**
   * Just to show this is here...
   Ptr<MinMaxAvgTotalCalculator<uint32_t> > test = 
   CreateObject<MinMaxAvgTotalCalculator<uint32_t> >();
   test->SetKey("test-dc");
   data.AddDataCalculator(test);

   test->Update(4);
   test->Update(8);
   test->Update(24);
   test->Update(12);
  **/

  // This DataCalculator connects directly to the transmit trace
  // provided by our Sender Application.  It records some basic
  // statistics about the sizes of the packets received (min, max,
  // avg, total # bytes), although in this scenaro they're fixed.
  Ptr<PacketSizeMinMaxAvgTotalCalculator> appTxPkts =
    CreateObject<PacketSizeMinMaxAvgTotalCalculator>();
  appTxPkts->SetKey ("tx-pkt-size");
  appTxPkts->SetContext ("node[0]");
  Config::Connect ("/NodeList/0/ApplicationList/*/$Sender/Tx",
                   MakeCallback
                     (&PacketSizeMinMaxAvgTotalCalculator::PacketUpdate,
                     appTxPkts));
  data.AddDataCalculator (appTxPkts);


  // Here we directly manipulate another DataCollector tracking min,
  // max, total, and average propagation delays.  Check out the Sender
  // and Receiver classes to see how packets are tagged with
  // timestamps to do this.
  Ptr<TimeMinMaxAvgTotalCalculator> delayStat =
    CreateObject<TimeMinMaxAvgTotalCalculator>();
  delayStat->SetKey ("delay");
  delayStat->SetContext (".");
  receiver->SetDelayTracker (delayStat);
  data.AddDataCalculator (delayStat);




  //------------------------------------------------------------
  //-- Run the simulation
  //--------------------------------------------
  NS_LOG_INFO ("Run Simulation.");
  Simulator::Run ();




  //------------------------------------------------------------
  //-- Generate statistics output.
  //--------------------------------------------

  // Pick an output writer based in the requested format.
  Ptr<DataOutputInterface> output = 0;
  if (format == "omnet") {
      NS_LOG_INFO ("Creating omnet formatted data output.");
      output = CreateObject<OmnetDataOutput>();
    } else if (format == "db") {
    #ifdef STATS_HAS_SQLITE3
      NS_LOG_INFO ("Creating sqlite formatted data output.");
      output = CreateObject<SqliteDataOutput>();
    #endif
    } else {
      NS_LOG_ERROR ("Unknown output format " << format);
    }

  // Finally, have that writer interrogate the DataCollector and save
  // the results.
  if (output != 0)
    output->Output (data);

  // Free any memory here at the end of this example.
  Simulator::Destroy ();

  // end main
}

一 給定本次仿真參數distance，format，experiment，strategy，runID在初始化的同時也可以通過命令行改變，這些參數用於從多次實驗中快速區分和組合數據。

二 創建節點和網絡模型

三 安裝協議棧，並分配IP

四 設置移動模型，這裏爲靜止，並給定初始位置

五 安裝應用，這裏安裝Sender / Receiver，自定義的見examples/stats/wifi-example-apps.h|cc

六 數據統計與收集，這是本文重點，下面具體分析。
這裏創建DataCollector對象來存儲運行信息，並通過Trace機制記錄收發端幀和分組傳輸情況。

1 記錄發端幀傳輸（基WIFI MAC對界）

通過CounterCalculator(src/stats/model/basic-data-calculators.h )類實現計數，利用Trace機制，當節點0上wifiNetDevice/Mac/MacTx變化（source），通過Config::Connect關聯，定義的TxCallback作爲sink函數調用，導致CounterCalculator::update調用即m_count++從而起到計數功能;

2 記錄收端幀傳輸（基WIFI MAC對界）

類似情況1，雖然這裏的sink函數是PacketConterCalculator::PacketUpdate（src/network/utils/packet-data-calculators.cc），但是該函數仍然是通過CounterCalculator::update實現計數，即利用Trace機制，當節點1上wifiNetDevice/Mac/MacRx變化（source），通過Config::Connect關聯;

3 記錄發端分組傳輸

也是通過PacketConterCalculator::PacketUpdate實現計數，利用Trace機制，當節點0上/Application/*/$Sender/Tx變化（source），通過通過Config::Connect關聯，定義的PacketConterCalculator::PacketUpdate作爲sink函數調用;

4 記錄收端分組接收

由於收端應用Receiver沒有定義traced source，故這裏沒有采用Trace機制，而是直接利用Receiver:;SetCounter直接操作，通過SetCounter顯示類型轉換，j將appRx賦值給Receiver內部計數器，從而實現計數

以上均是通過PacketConterCalculator(src/network/utils/packet-data-calculators.cc)或者CounterCalculator(src/stats/model/basic-data-calculators.h )實現傳輸單元的計數，\下一個將通過引入PacketSizeMinMaxAvgTotalCalculator (src/network/utils/packet-data-calculators.h|cc)和MinMaxAvgTotalCalculator(src/stats/model/basic-data-calculators.h)實現單元內大小的記錄。

5 記錄發端分組大小

這裏採用Trace機制，節點0上/Application/*/$Sender/Tx變化（source），通過通過Config::Connect關聯，定義的PacketSizeMinMaxAvgTotalCalculator::PacketUpdate作爲sink函數調用，從而MinMaxAvgTotalCalculator::Update實現大小的記錄。

6 記錄端到端產生分組時的延遲

類似情況4，不採用Trace機制，直接利用Receiver:;SetDelayTracker記錄傳世時延最值/平均值等

七 運行程序命令

八 統計結果輸出

對於輸出要麼OMNet++(純文本輸出格式)要麼SQLite(數據庫格式輸出)，這取決於程序頭部定義的參數format，並最終DataCollector對象進行存儲。

九 控制腳本實現最後運行

通過 一個簡單的控制腳本實現該仿真程序在不同距離下大量重複（作爲輸入）實驗後運行畫圖。可參考example/stats/wifi-example-db.sh（以後自己寫多個不同輸入下重複仿真項目時可參考這個）。該運行腳本每次都是基於一個不同的距離作爲輸入，收集每次仿真結果到SQLite數據庫，其中對於每個距離輸入，進行5次重複實驗以減小波動。全部仿真完成只需幾十秒，在完成存儲到數據庫後，可通過SQLite命令行進行SQL查詢。並調用 wifi-example.gnuplot畫圖

進入該目錄

產生data.db數據庫，wifi-default.data和wifi-default.eps圖

圖是對應距離下的丟包率以表徵WiFi模型性能。